Proc. Nat. Acad. Sci. USAVol. 10, No. 12, Part I, pp. 3669-3673, December 1973

Manipulation of Fatty Acid Composition in Animal Cells Grown in Culture*(LM cells/desthiobiotin/serum-free medium/membranes/detergents)

BERNADINE J. WISNIESKI, ROBERT E. WILLIAMS, AND C. FRED FOX

Department of Bacteriology and the Molecular Biology Institute, University of California, Los Angeles, Calif. 90024

Communicated by P. D. Boyer, July 3, 1973

ABSTRACT The fatty acid composition of animalcells cultured in serum-free medium can be manipulatedwhen the synthesis of endogenous fatty acids is inhibitedby a biotin analog and fatty acids are supplied in tbemedium as detergent esters of Tween. When mouse LMcells were grown in medium supplemented with Tween-19:0 (an ester ofTween and nonadecanoic acid), odd chainfatty acid content of cellular phospholipids and neutrallipids increased from 1% to 75%. Concurrently, the satu-rated fatty acid content increased from 27% to 85%. Simi-lar alterations in fatty acid content have been observedwhen BHK21 cells are subjected to the same enrichment re-gime. The ability to control the fatty acid composition ofcultured animal cells is a prerequisite to investigationsinto the role of the membrane lipid physical state in pro-cesses unique to these cells.

Microbial fatty acid auxotrophs have been used to modifythe fatty acid composition of biomembrane lipids in order tostudy the effect of lipid composition on membrane structure,function, and assembly. Research with fatty acid-requiringmutants of Escherichia coli has yielded significant insight intothe nature of membrane assembly (1, 2). These mutants havealso been used to demonstrate a dependence of transport rateon the physical state of membrane lipids (3-8).Animal cells have many functions with no counterpart in

microorganisms. They undergo oncogenic transformationand cell-cell fusion. Cell-cell fusion has been implicated in therelease of dormant tumor viruses (9). Membrane-envelopedviruses such as herpes, parainfluenza, influenza, and RNAtumor viruses infect by virtue of their ability to undergomembrane fusion with their host cells, or as a result of pino-cytosis (10). Some type of membrane alteration is implicatedin all these processes.

In order to manipulate the physical properties of animal cellmembranes, as has already been done in bacterial and fungalsystems (11-13), we have systematically searched for pro-cedures to force cultured animal cells to incorporate exog-enously-supplied fatty acids into membrane lipids. Bymanipulating the fatty acid composition of these cells, wecould analyze the effects of altered lipid fluidity on numerousmembrane functions, e.g., cell-cell fusion, virus-cell fusion,antigenic mixing, and phagocytosis.To date, no one has succeeded in grossly enriching the

phospholipids of cultured animal cells with a defined fatty

Abbreviations: Saturated, aliphatic, straight chain fatty acidsare designated 16: 0 for hexadecanoic acid, 17:0 for heptadecanoicacid, etc. Similarly, a fatty acid designated 16:1 is a 16 carbonstraight chain fatty acid containing one cis double bond.* A preliminary report of this work has been published [Wisni-eski, B. & Fox, C. F. (1973) Fed. Proc. 32, 556Abstr].

3669

acid supplied exogenously. The problems are as follows:(1) Growth media typically contain serum or other non-defined materials; many defined media contain Tween 80(fatty acid composition similar to olive oil) and cholesterol.(2) Unesterified (free) fatty acids supplied in ethanol or on analbumin carrier lead to lipid accumulation (steatosis) andirreversible cell degeneration (14, 15). (3) Endogenous fattyacid synthesis must be arrested, or alternatively, lipid-requiring mutants must be isolated. (4) Endogenous fattyacid desaturation must be controlled. By eliminating orminimizing these factors, we were able to enrich culturedanimal cells with fatty acids supplied exogenously, and toalter the degree of saturation in membrane lipid fatty acids.

MATERIALS AND METHODS

Organism. Mouse LM cells derived from NCTC clone 929(L cells) and adapted to growth on medium 199 plus 0.5%peptone were obtained from the American Type CultureCollection.

Media. Maintenance medium (MEM+P) was Eagle'sminimal essential medium with Earle's salts (medium MEM,GIBCO) plus 0.5% bacto-peptone (Difco). Chemically-defined medium (MEM+GV) consisted of MEM modifiedto contain twice the normal concentrations of glutamine andminimal essential vitamins. Medium MEM+GV (biotin-free) was supplemented with d-biotin (20 .ug/liter) to prepareMEM+GVB, with avidin (2 mg/liter, 13 units/mg) toprepare MEM+GVA, and with dl-desthiobiotin (2000,ug/liter) to prepare MEM+GVdB. All media contained 2.2g/liter of sodium bicarbonate, and were adjusted to pH 7.3with HCl before filtration.

Culture Procedures. Monolayer cultures were establishedin 250 ml Falcon flasks containing 15 ml of medium. Incuba-tion was at 370 with humidified air and 6% CO2. At con-fluency (approximately 108 cells per flask), cells were re-moved by rinsing the monolayer with 3 ml of 0.005% trypsin(Sigma: type 2, pancreatic; 1500 BAEE units/mg) dilutedwith Tris-saline (Tris, 3 g/liter; NaCl, 8 g/liter; Na2HPO4,0.1 g/liter; and 12 N HCl, 1.75 ml) at pH 7.4. Trypsin wasremoved immediately and the flask incubated for 5 min.Then, 3 ml of MEM was added and the monolayer dispersedby agitation. Aliquots of this cell suspension (0.1-0.25 ml)were used for inoculating the flasks. Medium was changedtwice weekly. When some cultures were first shifted to Tween-supplemented medium, they yielded viable, floating cells.These cells subsequently grew in monolayers when themedium containing them was added to 15 ml of MEM

Dow

nloa

ded

by g

uest

on

July

25,

202

0

3670 Cell Biology: Wisnieski et al.

f.4."~ ~ ~ ~

FIG. 1. Photomicrographs (X 100) of LM cells starved for

biotin by growth in MEM+G dB for 4 days and then shifted

to (A) MEM+P, (B) MEM+GVdB, (C) MEM+GVdB +

Tween-16 :0 (50 Ag/ml; approximately 10% 16:0 by weight), and

(D) MEM+G!dB + Tween-19:0 (100 Ig/ml; approximately

5% 19:0 by weight). Cultures were incubated in the specified

media for 7 days (Methods) prior to visual assessment of growth.

in a new flask. After cell attachment the MEM was replaced

with a Tween-supplemented medium.

Fatty Acid Supplements. Tween-acetate was synthesized

by reacting 5 g of Tween 20 with 20 ml of glacial acetic acid

in 100 ml of t-butyl alcohol plus 10 ml of 12 N HCl. The

reaction mixture was refluxed under N2 for 4 hr. Solventswere removed by vacuum distillation at 500.

Tween-fatty acid esters were synthesized by transesterifi-

cation of commercially available Tweens with excess fatty

acid (approximately 1:40 molar ratio) and also by trans-

esterification of Tween-acetate (see above) with a moderateexcess of fatty acid (1:2-1:5 molar ratio Tween-acetate to

fatty acid). A specific example of each method follows.

Method 1: Tween-palmitate (16:0) was prepared by reacting4 g of 16:0 and 1 g of Tween 40 in 25 ml of t-butyl alcoholplus 2.5 ml of 12 N HCl. The reaction mixture was refluxedunder N2 for 4 hr and solvents removed by vacuum evapora-

tion. Method 2: Tween-nonadecanoic acid (19:0) was pre-pared by refluxing 0.5 g of 19:0 and 1.0 g of Tween-acetatein 10 ml of t-butyl alcohol plus 1 ml 12 N HCL All fatty acids

were purchased from the Hormel Institute (Austin, Minn.)

and were more than 99% pure. Commercial grade Tweens are

heterogeneous with respect to the composition of the fattyacid moiety.

Tweens were separated from free fatty acids by column

chromatography. MN-silica gel, 70-325 mesh ASTM (E.

Merck Div.), was washed three times with glass distilledwater, and fine particles which remained in suspension were

removed after each wash. The gel was then filtered, washed

with methanol, refiltered and activated overnight at 1100.

Columns were packed with silica gel (10 g/g of sample) in

diethyl ether and washed with three column volumes ofether. Samples were applied in ether containing sufficientmethanol to solubilize the Tween portion of the sample.Fatty acids were eluted with four column volumes of ether andthe Tween was eluted with four column volumes of methanol.The purity of the Tween preparations was determined by

thin-layer chromatography and gas-liquid chromatography;absence of free fatty acids was verified by thin-layer chroma-tography and purity of fatty acid composition by gas-liquidchromatography. None of the Tweens contained detectablefree fatty acids. Tweens synthesized by Method 1 were atleast 90% pure in esterified fatty acid composition; Tweensmade by Method 2 were at least 98% pure. Methanol-freeTween preparations were solubilized in glass-distilled water(100 mg/ml) and autoclaved at 15 psi for 15 min.Lipid Extraction. Trypsinized cell suspensions were diluted

1: 10 with Tris-saline, pelleted by centrifugation (12,100 X gfor 15 min), washed once with Tris-saline and again pelleted.Cells were extracted with chloroform: methanol (2:1 v/v).The extract was filtered to remove debris, dried under N2,and suspended before application to predeveloped TLCplates (250 Am of silica gel G; Analtech). Chloroform: meth-anol: ammonium hydroxide (65: 35: 10 by volume) wasused to resolve the phospholipids. Tweens migrated with thesolvent front in this system.

Lipids were visualized by iodinating an exposed edge ofthe plate. Areas containing phosphatidylethanolamine andphosphatidylcholine were scraped and eluted with threevolumes of methanol followed by one volume of chloroform:methanol (2: 1). Lipids were transmethylated with BF3(14% in methanol).Gas-Liquid Chromatography. Gas-liquid chromatography of

methylated fatty acids was performed on a Perkin Elmer 990gas chromatograph. The analytical columns were 1/8 inch by6 ft stainless steel packed with 100/120 mesh (DMSC)Chromosorb W coated with 10% EGSS-X. Runs were tem-perature programmed from 1500 to 200° at 4°/min. Peakareas and retention times were calculated using an Info-tronics digital integrator.

RESULTSThe protocol developed to enrich LM cells with exogenously-supplied fatty acids involved growing the cells in modifiedEagle's minimal essential medium (see Methods) containingno serum, cholesterol, or extraneous lipids. Fatty acid-detergent esters (Tweens) which act as nontoxic carriers offatty acid supplements were chemically synthesized to con-tain a defined fatty acid moiety. To inhibit endogenous fattyacid synthesis, cells were starved for biotin in mediumMEM+GV ("biotin-free") containing avidin or desthio-biotin. Desthiobiotin treatment proved to be the more effec-tive method. Since fatty acid desaturase activity is sensitive tofatty acid chain length, the ratio of saturated to unsaturatedfatty acids in the cells was manipulated by using Tween-fattyacid esters of appropriate chain length.

Cultures maintained continuously in biotin-starvationmedium die within 1-2 weeks. However, the effects of a4-7 day starvation for biotin are reversible if cultures areshifted to growth in MEM + 0.5% peptone or to MEM+GVdB + Tween-fatty acid supplement (Fig. 1).

Cells shifted to Tween-19 :0 medium were analyzed forenrichment of 19:0 in phosphatidylcholine and phosphatidyl-

Proc. Nat. Acad. Sci. USA 70 (1973)

Dow

nloa

ded

by g

uest

on

July

25,

202

0

Manipulation of Fatty Acid Content in Animal Cells 3671

ethanolamine. These contain greater than 80% of the totalfatty acids in LM cell phospholipids (16). Odd chain fattyacids provided a marker for gas-liquid chromatography sincethe cells synthesize primarily even chain fatty acids. The fattyacid compositions of phospholipids from LM cells maintainedon MEM+P medium (0.3 Mg of biotin per g of peptone)and from cells shifted to MEM+GVdB supplemented withTween-19:0 are described in Fig. 2. Essentially no odd chainfatty acids were synthesized de novo in the presence of biotin(Fig. 2a). Cells in biotin-starvation medium containingTween-19:0 incorporated 19:0 into phospholipids at theexpense of even chain fatty acids (Fig. 2b). Biotin starvationis necessary to maximize the incorporation of exogenousfatty acid into phospholipid. In an experiment done in dupli-cate under conditions identical to those described for Fig. 2(MEM+GVdB + 100 Mg/ml of Tween-19:0), 33.4 and35.9% of the fatty acids in phosphatidylcholine plus phos-phatidylethanolamine were derived from Tween-19:0. Underconditions identical to those described for Fig. 2a (MEM+P),except for the addition of 100 gg/ml of Tween-19:0, only9.7 and 11.0% of the fatty acids in phosphatidylcholine plusphosphatidylethanolamine were derived from Tween-19:0.The predominant unsaturated fatty acid in the phospholipidsof control cultures is 18:1; the level of this fatty acid wasreduced in cells enriched for odd chain fatty acids. Althoughthere was always some desaturation of 19:0, the level ofdesaturation was greatly reduced compared to that of 18:0.Consequently, the saturated fatty acid content increasedproportionately to 19:0 enrichment. However, metabolicproducts of 19:0 (15:0, 17:0, 17:1, and 19:1) increasedwhen the period of growth with Tween-19:0 was lengthened(data not shown).Tables 1, 2, and 3 describe the fatty acid compositions of

TABLE 1. Fatty acid composition of phospholipids*extracted from control cultures

Culture mediumtMEM+GVB

MEM+ Mem+ MEM+ +Tw-P GVdB GVB acetate$

Fatty acid (%) (%) (%) (%)

19:1 0 0 0 019:0 0 0 0 018:1 59.8 56.7 45.9 46.818:0 10.0 9.9 12.7 18.217:1 0.6 0.2 0.3 017:0 0.1 0.1 0.3 0.916:1 9.4 8.6 7.6 5.816:0 15.8 18.7 27.3 21.915:0 0.5 0.8 0.8 014:0 1.4 2.0 3.0 3.312:0 0.2 0.4 0.3 0.2Other§ 2.2 2.6 1.8 2.9

% odd/% even 1.2/98.8 1.1/98.9 1.4/98.6 0.9/99.1%sat./% unsat. 28.0/72.0 31.9/68.1 44.4/56.4 44.5/56.5

* Phosphatidylcholine plus phosphatidylethanolamine.t See Methods for details; incubation periods for biotin star-

vation and for growth were as indicated for Fig. 1.t Tween-acetate concentration was 50.ug/ml.§ Fatty acids included in Other were even chain and un-

saturated. These consisted primarily of 20: 1.

TABLE 2. Fatty acid composition of phospholipids*extracted from Tween-19:0 cultures

Tween-19:0 concentration (jsg/ml)t

100 250 375 500Iog lMg og Jog

Fatty acid (%) (%) (%) (%)

19:1 2.0 7.1 5.8 0.519:0 13.6 19.9 33.4 72.918:1 36.3 35.9 28.6 11.918:0 8.7 4.1 5.8 2.617:1 1.4 3.0 3.2 0.417:0 1.0 0.5 1.3 1.016:1 10.2 11.5 6.5 2.716:0 17.4 12.2 11.3 5.315:0 1.4 1.2 0.9 0.414:0 5.1 3.5 2.7 1.512:0 1.6 0.1 0 0.8Othert 1.3 1.0 0.5 0

% odd/% even 19.4/80.6 31.7/68.3 44.6/55.4 75.2/24.8% sat./% unsat. 48.8/51.2 41.5/58.5 55.4/44.6 84.5/15.5

* Phosphatidylcholine plus phosphatidylethanolamine.t Tween-19:0 contained 5% 19:0 by weight. Procedures

were as described in Methods and Fig. 1.t Fatty acids included in Other were even chain and un-

saturated. These consisted primarily of 20: 1.

phospholipids extracted from control LM cultures (Table 1),and from LM cultures grown with various concentrations ofTween-19:0 (Table 2) and Tween-16:0 (Table 3). Medium

FIG. 2. Fatty acid composition of combined phosphatidyl-choline and phosphatidylethanolamine fractions from cellsgrown in (a) MEM+P and (b) MEM+GVdB + Tween-19:0(100 ;g/ml; approximately 5% 19:0 by weight). The time periodsfor initial biotin starvation (4 days) and for growth in the specifiedmedia (7 days) were as described in Fig. 1.

Proc. Nat. Acad. Sci. USA 70 (1973)

Dow

nloa

ded

by g

uest

on

July

25,

202

0

3672 Cell Biology: Wisnieski et al.

TABLE 3. Fatty acid composition of phospholipid&*extracted from Tween-16:0 cultures

Tween-16:0 concentration (ug/ml)t

50 100 200 250Jsg jg ug Mxg

Fatty acid (%) (%) (%) (%)

19:1 0 0 0 019:0 0 0 0 018:1 38.0 51.3 48.8 55.818:0 11.0 13.2 10.8 9.217:1 0.2 0.4 0.4 0.317:0 0.6 0.7 0.4 016:1 16.0 10.0 19.2 11.516:0 27.7 17.8 15.4 17.815:0 1.6 1.4 1.7 0.714:0 4.0 3.0 1.0 1.512:0 0 0 0.1 0Other$ 0.9 2.2 2.2 3.2

% odd/% even 2.4/97.6 2.5/97.5 2.5/97.5 1.0/99.0% sat./% unsat. 44.9/55.1 36.1/63.9 29.4/70.6 29.2/70.8

* Phosphatidylcholine plus phosphatidylethanolamine.t Tween-16:0 contained 10% 16:0 by weight. Procedures

were as described in Methods and Fig. 1.t Fatty acids included in Other were even chain and un-

saturated. These consisted primarily of 20: 1.

MEM+P was used to maintain stock cultures; MEM+GVdB was used for biotin starvation and was subsequentlysupplemented with Tween-fatty acid esters to effect enrich-ment. Biotin supplementation of MEM+GV medium (i.e.,MEM+GVB) did not suffice to maintain cell growth in-definitely, nor did it replace Tween-fatty acid esters in pro-moting the growth of biotin-starved cells. Our present infor-mation cannot account for this phenomenon.We can manipulate the degree of enrichment for an exog-

enously supplied fatty acid by increasing the growth time inmedium supplemented with a Tween-fatty acid ester suppliedat constant molarity, or by increasing the concentration of- theTween-fatty acid ester supplement. The extent to which wecan enrich cells for 19:0 is demonstrated in Table 2. Anincrease in the concentration of the fatty acid supplementeffected increases in enrichment for both odd chain and satu-rated fatty acids. When a high concentration of Tween-19 :0was supplied (500 Mug of Tween-19:0 per ml), 75% of the fattyacids in phospholipids were derived from 19:0. Enrichmentin 19:0 alone was 73%. The saturated fatty acid content in-creased from 27% in cells grown in MEM+P to 85% incells grown with 500 MAg/ml of Tween-19:0. The neutral lipidfractions derived from these cells had a fatty acid contentsimilar to that of the phospholipids (data not shown). Celllysis was encountered in medium supplemented with higherconcentrations of Tween-19: 0 (approximately 750 Mug/mI).Although we have not succeeded in totally blocking endoge-nous fatty acid synthesis, high (:oncentrations of Tween-19: 0apparently inhibit a portion of this background synthesis.The data in Table 3 indicate that cells apparently have

an increased capacity for elongating 16:0 to C18 fatty acidswhen the concentration of 16:0 supplement is increased.Whatever the reason for this, cells grown in medium contain-

cellular growth (approximately 250 ug of Tween-16:0 per ml)have a fatty acid profile similar to that of cells grown incontrol cultures (Table 1).BHK21 cells adapted to growth in serum-free medium

(MEM+P) and starved for biotin in MEM+GVdB havealso been enriched for 19:0 after growth in MEM+GVdBsupplemented with Tween-19:0. The protocol we haveestablished for enriching LM cells with specific fatty acidsmay be applicable to other cell lines as well.

DISCUSSION

We have established a protocol for enriching phospholipidsof animal cells grown in culture with fatty acids suppliedexogenously. Cells are first starved for biotin in a lipid-freeminimal medium until growth is inhibited. The medium issubsequently supplemented with a fatty acid esterified toTween and growth resumes.

Incorporation of exogenous fatty acids into cellular phos-pholipids was initially verified with a radioactive fatty acidesterified to Tween. To verify enrichment, Tween-19:0 (anester of Tween and nonadecanoic acid) was employed as amedium supplement. This supplement was used since thecells can not synthesize a significant amount of odd chainfatty acids, and since odd and even chain fatty acids and theirdesaturation products are easily resolved by gas-liquidchromatography. Odd chain fatty acids were incorporated intoboth phospholipids and neutral lipids when cells were grownin medium containing a Tween-19:0 supplement. The fattyacid compositions of both these fractions contained as muchas 73% 19:0. The results from published in vitro studies pre-dicted that desaturation of 19:0 would be minimal (17, 18).As expected, the proportion of saturated fatty acids wasgreatly increased, from 27% in controls, to 85% in 19:0-enriched cells.

Experiments with fatty acid auxotrophs of E. coli havedemonstrated that parallel changes occur in transport activityand membrane lipid fluidity at characteristic temperaturesdetermined by the fatty acid composition of membrane phos-pholipids (3-8). The characteristic temperatures for transportcorrelate with characteristic temperatures defining the endpoints (upper and lower) of the course of lateral phase sepa-rations in membrane lipids as monitored by electron spinresonance spectroscopy (8, 19) and x-ray diffraction (Blasie,K., Linden, C. D. & Fox, C. F., to be published). In addition,the lateral mobility of membrane lipids is apparently arrestedbelow the lower characteristic temperature (tj) (ref. 2;Hubbell, W., unpublished observations), and at least somemembrane assembly processes are abortive below t (1).Membranes with a high saturated fatty acid content have a

high tj and membranes with a low saturated fatty acidcontent, a low tI. Since we can now control the ratio of satu-rated to unsaturated fatty acids in established lines of animalcells (Tables 1, 2, and 3), we hope to determine how altera-tions in ti, and thus in membrane lipid fluidity, affect mem-brane functions unique to these cells.

Alterations in animal cell membranes are implicated inthe etiology of many disease states and in aging. Mem-branes are targets for viruses and bacteria, and for certainhormones and drugs. Feedback between the cellular genomeand membranes appears to regulate density-dependent cellgrowth (i.e., contact inhibition). Cell-cell and virus-cellfusion are certainly membrane phenomena. A wealth of in-ing the highest concentration of Tween-16: 0 compatible with

Proc. Nat. Acad. Sci. USA 70 (1973)

Dow

nloa

ded

by g

uest

on

July

25,

202

0

Manipulation of Fatty Acid Content in Animal Cells 3673

formation exists linking membrane alterations with theseprocesses and others, such as oncogenic transformation.The protocol we have developed enables us to take novel

approaches to studying these problems. Our ability to manip-ulate the fatty acid composition of BHK21 cells may allow usto investigate possible interrelationships between membranefluidity and the rate and/or degree of oncogenic transforma-tion. The lipid compositions of cells and enveloped viruscan be varied independently by harvesting virus from cellswith a fatty acid composition different from that of the ex-perimental host cells. It should thus be possible to assessindependently the role of cellular and viral membrane fluidityin the infiltration of cells by enveloped viruses.

Since some membrane-associated enzymes are sensitive tomembrane phase behavior (6, 20, 21), we should be able tomonitor fatty acid enrichment in several different membranefractions (e.g., mitochondrial, plasma, etc.) and proceed todetermine how lipid fluidity affects enzyme activity and regu-latory phenomena. The known lipid dependence of adenylcyclase activity is of considerable interest here (22-25).Immunological response may also be affected by membranelipid fluidity. The clustering of antigen receptors on lympho-cytes is apparently a prerequisite to subsequent mitogenicresponse (26-28). This clustering undoubtedly requires mem-brane fluidity.

This work was supported by Research Grants GM-18233 andAI-17033 from the United States Public Health Service, GrantDRG-1153 from the Damon Runyon Fund, and Grant BC-79from the American Cancer Society. C.F.F. is the recipient ofUnited States Public Health Service Career Development AwardGM-42359; B.J.W. is a Postdoctoral Fellow of the DamonRunyon Fund (DRF-749).

1. Tsukagoshi, N. & Fox, C. F. (1973) Biochemistry 12, 2816-2822.

2. Tsukagoshi, N. & Fox, C. F. (1973) Biochemistry 12, 2822-2829.

3. Schairer, H. U. & Overath, P (1969) J. Mol. Biol. 44, 209-214.

4. Overath, P., Schairer, H. U. & Stoffel, W. (1970) Proc.Nat. Acad. Sci. USA 67, 606-612.

5. Wilson, G., Rose, S. P. & Fox, C. F. (1970) Biochem.Biophys. Res. Commun. 38, 617-623.

6. Esfahani, M., Limbrick, A. R., Knutton, S., Oka, T. &Wakil, S. J. (1971) Proc. Nat. Acad. Sci. USA 68, 3180-3184.

7. Wilson, G. &Fox, C. F. (1971) J. Mol. Biol. 55, 49-60.8. Linden, C. D., Wright, K. L., McConnell, H. M. & Fox,

C. F. (1973) Proc. Nat. Acad. Sci. USA 70, 2271-2275.9. Watkins, J. F. & Dulbecco, R. (1967) Proc. Nat. Acad. Sci.

USA 58, 1396-1403.10. Morgan, C., Rose, H. M. & Mednis, B. (1968) J. Virol. 2,

507-516.11. Machtiger, N. A. & Fox, C. F. (1973) Annu. Rev. Biochem.

42, 507-516.12. Keith, A. D., Wisnieski, B. J., Williams, J. C. & Henry, S.

(1973) in Lipids and Biomembranes of Eukaryotic Micro-organisms, ed. Erwin, J. A. (Academic Press, New York),pp. 259-321.

13. Brody, S. & Allen, B. (1972) J. Supramolecular Struct. 1,125-134.

14. Geyer, R. P. (1967) in Lipid Metabolism in Tissue CultureCells, eds. Rothblat, G. H. & Kritchevsky, D. (The WistarInstitute Press, Philadelphia, Pa.), pp. 33-47.

15. Moskowitz, M. S. (1967) in Lipid Metabolism in TissueCulture Cells, ed. Rothblat, G. H. & Kritchevsky, D. (TheWistar Institute Press, Philadelphia, Pa.), pp. 49-62.

16. Anderson, R. E., Cumming, R. B., Walton, M. & Snyder, F.(1969) Biochim. Biophys. Acta 176, 491-501.

17. Morris, L. J. (1970) Biochem. J. 118, 681-693.18. Brett, D., Howling, 1)., Morris, L. J. & James, A. T. (1971)

Arch. Biochem. Biophys. 143, 535-547.19. Linden, C. D., Keith, A. D. & Fox, C. F. (1973) J. Supra-

molecular Struct., in press.20. Esfahani, M., Crowfoot, P. D. & Wakil, S. J. (1972) J. Biol.

Chem. 247, 7251-7256.21. Mavis, R. D. & Vagelos, P. R. (1972) J. Biol. Chem. 247,

652-659.22. Rethy, A., Tomasi, V. & Trevisani, A. (1971) Arch. Biochem.

Biophys. 147,36-40.23. Pohl, S. L., Krans, H. M. J., Kozyreff, V., Birnbaumer, L. &

Rodbell, M. (1971) J. Biol. Chem. 246, 4447-4454.24. Levey, G. S. (1971) Biochem. Biophys. Res. Commun. 43,

108-113.25. Rethy, A., Tomasi, Y., Trevisani, A. & Barnabei, 0. (1972)

Biochim. Biophys. Acta 290, 58-69.26. Taylor, R. B., Duffus, W. P. H., Raff, M. C. & de Petris, S.

(1971) Nature New Biol. 233, 225-229.27. Yahara, I. & Edelman, G. M. (1972) Proc. Nat. Acad. Sci.

USA 69, 608-612.28. Edelman, G. M., Yahara, I. & Wang, J. I. (1973) Proc.

Nat. Acad. Sci. USA 70, 1442-1446.

Proc. Nat, Acad. Sci. USA 70 (1973)

Dow

nloa

ded

by g

uest

on

July

25,

202

0